JP3327420B2 - Postal code search method by simple address input - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a postal code search method by simple postal address input for efficiently performing a postal code search and writing a postal code when a postal code is not described in a postal matter.

[0002]

2. Description of the Related Art When an individual issues a postal matter, after writing an address on an envelope or postcard, the postal code corresponding to the address is manually entered by referring to a postal guide (postal code book). I searched and entered it in the postal code box on the envelope or postcard. According to the Postal Guide, the prefectures are listed in the order from Hokkaido to Okinawa.
907 postal code has been determined and the city, ward, county,
Postal codes are determined for towns and villages in alphabetical order. Therefore, when searching for a postal code with a postal guide,
First, referring to the page of the prefecture in the table of contents, and then searching for the city, ward, county, town, or village name in the prefecture, the number shown on the right side of the corresponding item is the postal code to be obtained. Also,
If the number is not described on the right side of the city, ward, town, or village name, the number on the right side where “other than above ...” is displayed is the desired postal code. However, when mail is issued by a company or an organization, there is often no time to manually search for a postal guide because the volume is enormous. In such a case, an automatic search is performed by a computer to find the corresponding postal code, and the postal code is entered in the entry field of each mail, or stamped with a rubber stamp. In that case, generally, in order to search for a postal code from a character string representing an address (hereinafter, referred to as an address string), it was necessary to input the full text of the address from an input device of a computer. In addition, a method has been proposed in which a full address string and a postal code are searched using a town name as a keyword as a part of an address (see, for example, JP-A-61-75468). In this method, the entry of a place name is made more efficient by registering a postal code, a prefecture name, and a municipal name in a memory using a town name as a keyword. That is, since the words and phrases other than the municipalities are common to the respective municipalities, it is determined that it is inappropriate to use them as keywords. For example, if "Kasugaseki" is input and converted, "Zip code 100, Kasumigaseki, Chiyoda-ku, Tokyo" is displayed on the screen, so that the time for the search operation can be reduced.

[0003]

The conventional search method, that is, a method of searching for a postal code by inputting the whole or most of the address string, has the advantage that the postal code is always determined to be one. However, it takes a lot of time and effort to input a full text, which is extremely inefficient. In addition, the above-mentioned proposed method is simple because it is only necessary to input the name of a town or village, the name of a capital, etc. It was difficult to narrow down the postal code to one. SUMMARY OF THE INVENTION An object of the present invention is to solve such a conventional problem and to provide a postal code search method by simple address input that can uniquely determine a postal code with a short number of input characters without inputting the entire text of an address string. Is to provide.

[0004]

In order to achieve the above object, a method for searching for a postal code by simple address input according to the present invention comprises the steps of: (a) dividing an address into a plurality of layers and selecting a part of the layers;
Then, the address string of the part of the hierarchy is input to the calculator, and is omitted.
Address column was the a zip code search how to search for a postal code that corresponds to the original, as the omitted address column, the address column,
A first level consisting of prefecture names, a second level consisting of city / county names and ward names in Tokyo, a third level consisting of ward names / town / village names, and a fourth level consisting of capital letters / city names. Is divided into a fifth hierarchy composed of common names such as large-sized and inter-character spaces, a sixth hierarchy composed of character names / chome, and a seventh hierarchy composed of small-letter names. Two or more hierarchies are selected in advance, a partial or all character strings of the selected hierarchies are used, and the character strings are used as keywords to search for postal codes, respectively. It is characterized in that the number is the final result. (B) As the abbreviated address column, the second hierarchy (name of city / county / name in ward) and the fourth hierarchy (name of capital / city / town) of the above first to seventh hierarchy are designated. , Or the first character of one kanji and the entire character string of the other, and the postal code corresponding to the address string is stored for each hierarchy using the first character or the entire character string as a keyword. Each database is searched, and a common postal code in each search result is set as a final result.

[0005]

According to the present invention, the address string is classified into several levels such as a prefecture name, a city name, and a town name, and a character string for searching for a new postal code using part or all of the address for each level. (Keyword). By performing a postal code search using this character string, the efficiency of address input is improved. In addition, by creating a table in which a new character string (keyword) is associated with a corresponding postal code, a search method with high input efficiency is realized. In particular,
The address string is divided into sections from a first layer (prefecture name) to a seventh layer (small letter name), and two or more layers designated in advance, here, the most characteristic layer in the address string. Designate two levels (city / county name / Tokyo ward name) and fourth level (Oaza name / city / town name), the first letter of both levels, or the first letter of one level and all of the other level Select a string and
A search is performed using the first character or the entire character string as a keyword. Then, of the postal code candidates obtained in each search, the postal code common to both layers is determined as the postal code of the input keyword. Alternatively, using the first character or the entire character string as a keyword, the candidate postal codes are sequentially narrowed down to obtain a final result. As described above, in the present invention, the postal code can be determined based on the short abbreviated address string, so that the postal code can be added to a large amount of mail in a short time, and the efficiency of the mailing operation can be improved. it can.

[0006]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is an explanatory diagram showing an example of hierarchical classification of an address string in the present invention. FIG. 1 shows an example in which address strings are classified into seven layers as a method of layer classification.
That is, here, the highest level (first level) of the seven levels
And the name of the prefecture, the name of the city / county / ward in the next level (second level), the name of the ward / town in the next level (third level), The name of the Oaza / city / town name as the hierarchy, the common name (the Oaza / character space) as the next hierarchy (fifth hierarchy), and the character name / chome as the next hierarchy (sixth hierarchy). The small font names are classified as the hierarchy (seventh hierarchy). For example, if "zip code 100-1-16 Kasumigaseki, Chiyoda-ku, Tokyo" is extracted for each classification, "Tokyo" belongs to the first hierarchy, "Chiyoda-ku" belongs to the second hierarchy, and "Kasumigaseki ] Belongs to the name of the city and town on the fourth level, and "1.chome" belongs to the name of the sixth level.

In the present invention, for example, one kanji character on the second level (city name) and all the characters on the fourth level (large name / city name) are input from the input device. Enter it. In other words, as can be seen from the above example of "Kasumigaseki", paying attention to the fact that the city and county name on the second level and the Oaza name and city / town name on the fourth level are the most characteristic parts in the address, I thought that it would be possible to search for a postal code just by entering these two sets without having to enter all address names. Further, by extracting a common postal code using these two sets of search results, a high-speed and accurate postal code can be obtained. Note that all characters in the second and fourth hierarchies may be used as keywords, but it is sufficient that only one character is used as a keyword in one of them, and that both use only one character as a keyword. Turned out to be. Here, an example will be described in which the first character of the second hierarchy and all the characters of the fourth hierarchy are used as keywords. Now, when the address is "Ochi-cho, Tokudani, Nagakute-cho, Aichi-gun, Aichi-gun," we focus on "Aichi-gun" on the second level and "Naga-chu" on the fourth level. That is, the first character “Ai” of the second hierarchy “Aichi” and all the characters of the fourth hierarchy “Nagachu” are input. A search is performed using these two terms as keywords.

FIG. 2 is an explanatory diagram showing an example of a postal code search using a keyword according to the present invention. In FIG. 2, 1
0 is a database in which the correspondence between the city / county / city ward name and the postal code in the second level is stored, and 11 is the capital letter / fourth level in the fourth level.
This is a database storing correspondence between city names and postal codes. These databases 10 and 11 are a database to which a corresponding zip code is assigned in advance corresponding to the name of a city or a county or a ward in Tokyo, and a database to which a corresponding zip code is assigned corresponding to a capital letter or a name of a city or town. Create it. Now, FIG. 2 shows a method of searching for a postal code using “Ai” and “Nagachu” as keywords. The search conditions are a forward match search and a search for each address notation in each hierarchy. According to the example of FIG. 2, first, a search for "love" is performed from the database 10 of one kanji character on the second layer, and the postal code as a candidate is determined. In FIG. 2, three candidates 243-03, 257, and 480-11 have been extracted, and thus cannot be determined. Next, a search for "Nagachu" is performed from the database 11 of all characters in the fourth hierarchy, and a postal code as a candidate is also searched. In FIG. 2, since the number of 480-11 candidates was reduced to one, the number was determined. Finally, a postal code common to these two sets of candidates is selected and determined as the postal code of the input keyword. In addition, in the case of FIG. 2, even if several candidates were obtained by searching the database 10, two common candidates were inevitably determined because the database 11 was narrowed down to one candidate. . However, when a plurality of candidates are obtained by searching the database 10 and a plurality of candidates are obtained by searching the database 11, it is important to extract a common postal code from these two sets of candidates. . Also, instead of searching for postal codes for each layer, a method of narrowing down the postal codes sequentially from the upper layer and obtaining a final result is also possible.
In the present embodiment, a method has been described in which a postal code is searched for each layer and a common one is obtained from them, but the same applies to a narrowing down method. Also, even if the postal code is not finally determined to be one, the postal code candidates are sufficiently narrowed down, so that the postal code can be easily selected manually.

FIG. 3 is a view showing the result of a postal code search by inputting a keyword address. The graph of FIG. 3 shows the result of evaluation based on the national street character file of the Geographical Survey Institute of Japan by the search method of the present invention. The horizontal axis shows the number of narrowed postal code candidates as the postal code search result, and the vertical axis shows the cumulative ratio to be determined. According to the results of FIG. 3, the probability that the postal code is narrowed down to one, that is, the ratio of uniquely determining the postal code is 90%, and the ratio of the postal code candidate number being determined to three or less is approximately 99%. The percentage of four or less or five or less is almost 100%. As described above, according to the present invention, the case of 90% is determined to be one postal code, so that the postal code can be searched in a short time. In the present embodiment, the kanji 1 of the second hierarchy is used as the keyword for the postal code search.
Although the example using the first character and all the characters in the fourth hierarchy has been described, the present invention is not limited to this.
For example, the first kanji character “Ai” in the second hierarchy and the first kanji character “Length” in the fourth hierarchy can be used as keywords.

[0010]

As described above, according to the present invention,
When searching for a postal code using a computer, the postal code can be uniquely determined with a short number of input characters. Efficiency can be improved.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a hierarchical classification of an address string according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing an example of a postal code search using a keyword in the present invention.

FIG. 3 is a diagram showing a search result of a postal code by inputting a keyword address according to the present invention;

[Explanation of symbols]

10 Database of postal code storage by second-level address 11 Database of postal code storage by fourth-level address

────────────────────────────────────────────────── ─── Continuing from the front page (72) Inventor Izawa Ichiro 1-6-119 Azabudai, Minato-ku, Tokyo Inside the Research and Development Center, Post and Telecommunications Research Institute, Ministry of Posts and Telecommunications Address: Hitachi Systems, Ltd. Office Systems Division (72) Inventor Atsushi Kato 1 Ikegami, Haruoka-cho, Owariasahi-shi, Aichi Pref., Japan Office Systems Division, Hitachi Ltd. (56) References JP-A-59-188737 (JP, A) JP-A-4-3699764 (JP, A) JP-A-2-309450 (JP, A) JP-A-5-225238 (JP, A) FACOM OS IV / X8 FSP KUIN Online Manual, Japan, Fujitsu Co., Ltd., September 30, 1983, pp. 1 to 37 (58) Fields investigated (Int.Cl. ⁷ , DB name) G06F 17 / 30 JICST file (JOIS)

Claims (2)

(57) [Claims] 1. An address is divided into a plurality of layers, and a part of the
And input the address string of the part of the hierarchy into the calculator,
Abbreviated address column a zip code search how to search for a postal code that corresponds to the original, as the omitted address column, the address column, and the first layer consisting of a prefecture name, cities and counties name, Tokyo The second level consisting of the ward name, the third level consisting of the ward name and the name of the town in the county, the fourth level consisting of the Oaza name and the city name, and the fifth level consisting of the common names such as the Oaza and inter-block areas And a sixth layer consisting of character names / chome and a seventh layer consisting of small names, and two or more layers among the first to seventh layers are selected in advance, and the selected layer is selected. Using a partial address string or a whole character string, searching for a postal code using each of the character strings as a keyword, and using a common postal code in each search result as a final result. Postal code search method. 2. The postal code search method according to claim 1, wherein the abbreviated address string is a second one of the first to seventh hierarchies (name of a city, county, or city). And the fourth hierarchy (name of the capital and city name), select the first kanji character of both hierarchies, or the first kanji first character and the entire other character string, and select the first character or the entire character string. A postal code by simple address input, wherein a database in which postal codes corresponding to an address sequence are stored for each hierarchy as a keyword is searched, and a common postal code in each search result is set as a final result. Number search method.